3WJ RNA Nanoparticles-Aptamer Functionalized Exosomes From M2 Macrophages Target BMSCs to Promote the Healing of Bone Fractures

Abstract

Up to now, impaired bone regeneration severely affects the healing of bone fractures, thus bringing tremendous suffering to patients. As a vital mediator between inflammatory response and bone regeneration, M2 macrophage-derived exosomes (M2-Exos) attenuate inflammation and promote tissue repair. However, due to a lack of specific targeting property, M2-Exos will be rapidly eliminated after systematic administration, thus compromising their effectiveness in promoting bone regeneration. To solve this hurdle, we initially harvested and characterized the pro-osteogenic properties of M2-Exos. A bone marrow mesenchymal stem cell (BMSC)-specific aptamer was synthesized and 3-way junction (3WJ) RNA nanoparticles were applied to conjugate the BMSC-specific aptamer and M2-Exos. In vitro assays revealed that M2-Exos bore the representative features of exosomes and significantly promoted the proliferation, migration, and osteogenic differentiation of BMSCs. 3WJ RNA nanoparticles-aptamer functionalized M2-Exos (3WJ-BMSC_apt/M2-Exos) maintained the original physical characteristics of M2-Exos, but bore a high specific binding ability to BMSCs. Furthermore, when being systemically administered in the mice model with femoral bone fractures, these functionalized M2-Exos mainly accumulated at the bone fracture site with a slow release of exosomal cargo, thereby significantly accelerating the healing processes compared with the M2-Exos group. Our study indicated that the 3WJ-BMSC_apt/M2-Exos with BMSCs targeting ability and controlled release would be a promising strategy to treat bone fractures.

Graphical Abstract

Graphical Abstract.

Significance Statement.

Extensive studies have confirmed that M2 macrophage-derived exosomes (M2-Exos) exhibit excellent osteogenic properties. However, M2-Exos have no specific targeting property. To solve this hurdle, aptamer-functionalized M2-Exos were designed, fabricated, and delivered by 3WJ RNA nanoparticles to target bone fracture. Our results demonstrated that the aptamer-functionalized M2-Exos can not only target bone marrow mesenchymal stem cells in vitro but also the fracture site after systemically intravenous injection. Moreover, they significantly accelerated the healing of femur bone fracture compared with the control groups. Our study validated that the specific targeting ability of M2-Exos could enhance bone regeneration, which holds great application potential in treating bone fracture.

Introduction

Bone fracture is a common disease worldwide, which creates a huge financial burden on society.^1,2 Globally, there were 178 million new cases and 445 million cumulative cases in 2019, which will increase concomitantly with the aging of the population.³ Despite the inherent ability of bone to regenerate and current advances in the treatment, approximately 5%-10% of patients suffer from poor healing of bone fracture, resulting in delayed union or nonunion, thus bringing significant stress and psychosocial disruption to society and patients.^4-6 Autologous bone grafting has long been used to treat delayed union or nonunion fracture. However, the limited supply restricts the clinical application.^7-9 Scientists have explored some alternative strategies to accelerate bone regeneration. For example, the incorporation of bone morphogenetic proteins (BMPs) confers synthetic bone graft osteoinductivity. Nevertheless, the usage of BMPs raises serious concerns due to rapid release and the resultant high concentration may lead to an increased risk of cancer, local inflammatory reactions, and heterotopic bone formation.^7,10-12 Therefore, it is urgent to explore a safe, effective, and innovative treatment for bone fracture.

Successful bone fracture healing requires a coordinated interaction between macrophages and bone marrow mesenchymal stem cells (BMSCs). During the bone repair process, macrophage polarization plays a key role in regulating the differentiation of BMSCs.¹³ M1 macrophages mainly play a role to induce an immune response, while M2 macrophages mainly participate in the tissue repair of bone fracture by promoting the osteogenesis of BMSCs.¹³ Meanwhile, M2 macrophages have been gradually recognized as a positive regulator in bone fracture healing.¹⁴ Recent studies have shown that macrophages promote bone fracture healing through paracrine mechanisms and exosomes are the main mediators of paracrine action.^15,16 Exosomes, approximately 40-160 nm in diameter, are extracellular vesicles with phospholipid bilayers secreted by almost all cell types.¹⁷ It contains a variety of components of the source cell, including miRNAs, lipids, cellular metabolites, and cell surface proteins.¹⁸ These specific components within the exosomes can carry and deliver essential signaling molecules to the recipient cells, which play a vital role in intercellular communication, thus influencing the physiological and pathological state of target cells.¹⁹ Recent studies have shown that M2-Exos are rich in miR-690, and miR-378a-3p, and could regulate osteogenic differentiation.^15,20 Although topical application of M2 macrophage-derived exosomes (M2-Exos) could accelerate the healing of femur fracture in mice and the repair of cranial defects in rats,^21,22 the role of systemic administration of exosomes for bone fracture healing is unpredicted due to no specific targeting property. Besides, unmodified exosomes are rapidly accumulated in and removed by reticuloendothelial organs, such as the liver and spleen.^23,24 Therefore, site-specific targeted exosomes need to be designed and fabricated to meet the requirement of systemic administration for bone fracture therapy.

RNA nanotechnology was first demonstrated in 1998 and has been a hotspot recently.²⁵ 3WJ RNA nanoparticles, obtained by bottom-up self-assembly of the packaging RNA (pRNA) in the bacteriophage phi29 DNA transport motor, have shown great potential as a new generation of targeted tumor therapeutic agents in the medical field.^26,27 The thermostability property allows 3-way junction (3WJ) to carry therapeutic fragments, including siRNA, nucleases, and aptamers, and retains all true folding and independent functional modules, allowing the preparation of highly stable 3WJ RNA nanoparticles.²⁸ Furthermore, 2ʹ-Fluoro (2ʹ-F) or 2ʹ-O-methyl (2ʹ-OMe) modification can lead to a significant increase in the thermal stability of 3WJ RNA nanoparticles and prevent their degradation by RNases.²⁹ Concomitantly, the 3WJ motif can simultaneously combine an imaging module, a targeting module, and a therapeutic module that has been widely used for targeted therapy for diverse tumors.^30-34 However, few studies use engineered exosomes synthesized by RNA nanotechnology for targeting and treating bone fracture.

Here, we aim to employ 3WJ RNA nanoparticles conjugated BMSC aptamers to enhance the targeting ability of M2-Exos and thus accelerate the healing of bone fracture in mice. The targeting property was achieved through previously reported BMSC aptamer sequences.^35,36 3WJ branch-linked cholesterol was used to bind to M2-Exos as a therapeutic module. Meanwhile, another branch of the 3WJ RNA nanoparticle was attached to Alexa 647 as an imaging module. Flow cytometry and confocal microscopy analysis were employed to monitor the targeting ability of 3WJ RNA nanoparticles-conjugated M2-Exos (3WJ-BMSC_apt/M2-Exos) to BMSCs in vitro. After that, a mouse femur fracture model was established and 3WJ-BMSC_apt/M2-Exos were systematically administered through tail vein injection. The bone fracture healing was measured by clinical and radiographic examinations. Based on the above results, we identified 3WJ-BMSC_apt/M2-Exos as an effective strategy for accelerating bone fracture healing.

Material and Methods

Cell Culture

Mouse macrophage line RAW264.7 was purchased from Procell and cultured in high glucose Dulbecco’s modified Eagle’s medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS; Gibco), 100 U/mL penicillin and streptomycin (Gibco). Mouse BMSCs were also purchased from Procell and cultured using a mouse mesenchymal stem cell complete medium. To obtain M2 macrophages, the RAW264.7 cells were induced by the complete culture medium with the addition of 20 ng/mL IL-4 (PeproTech, USA) for 24 h. To induce osteogenesis, BMSCs were cultured in osteogenic media containing 50 nM ascorbic acid, 10 mM β-glycerophosphate, and 100 nM dexamethasone. All experimental cells were incubated at 37°C in an incubator containing 5% CO₂.

Identification of M2 Macrophages

The induced RAW264.7 cells were digested and adjusted to the concentration of 10 × 10⁷/mL. A 100 μL of the cell suspension was applied for subsequent processing. Flow cytometric antibodies were both purchased from Biolegend (San Diego). FITC-labeled anti-mouse F4/80 and APC-labeled anti-mouse CD206 were used to identify M2 macrophages. CD206 is an intracellular protein and requires breaking the membrane using Cyto-Fast FIX/Perm Buffer (BD Biosciences) before adding flow cytometric antibodies for 30 minutes incubation at 4°C. Finally, flow cytometry was used for M2 macrophage identification and analysis.

The Isolation and Characterization of Exosomes

First, FBS was centrifuged at 120 000g for 16 h at 4°C to remove exosomes. Every 5.5-6 × 10⁶ successfully induced M2 macrophage was cultured in a 175 cm² cell culture flask (Corning) for 48 h, and 20 mL supernatant was collected. Dead cells and cell debris were removed by 2-step centrifugation at 2 930g for 10 minutes and 10 000g for 30 minutes at 4°C. After centrifugation, the filtrate was transferred to a new centrifuge tube over a 0.22 μm filter (Millipore, Billerica). The collected filtrate was centrifuged in a Type Ti90 rotor using an optima XPN-100 ultracentrifuge (Beckman Coulter) at 150 000g for 90 minutes at 4°C and resuspended using phosphate-buffered saline (PBS;). Then, the ultra-filtered exosomes were ultracentrifuged again at 150 000g, 4°C for 90 minutes, and a white precipitate, the exosomes, could be seen on the wall of the tube after the removal of the supernatant. The exosomes were re-suspended using a small amount of PBS and placed in a −80°C refrigerator for subsequent experiments.

Next, exosomal protein concentration was quantified using the BCA protein assay kit (Thermo Fisher Scientific). Exosomes were added dropwise to a carrier copper grid for 30 minutes and then negatively stained with uranium peroxide dropwise for 10 min at room temperature, excess stain was blotted off on filter paper and left to dry for 30 min. A JEM-1400PLUS transmission electron microscopy (TEM; JEOL, Japan) was employed to characterize exosomal morphology. Nanoparticle tracking analysis (NTA) was applied by a nanoparticle tracking analyzer (Particle Metrix GmbH, Germany) to analyze the distribution and concentration of particle size. Samples were diluted at the appropriate concentration and injected uniformly into the detection pipeline. Western blot (WB) was performed to detect specific positive exosome markers such as anti-Alix, anti-Flot, and anti-HSP70, and a specific negative exosome marker, Calreticulin (1:1000, Abcam, Cambridge, UK).

Fluorescent Labeling and Uptake of Exosomes

The membrane fluorescent dye DiO was added to the exosomes at a concentration of 10 μM/L and incubated for 30 min at 37 °C, and the free DiO was removed using Exosome Spin Columns (Life Technologies). The DiO fluorescently labeled exosomes (50 μg/mL) were incubated with BMSCs for 4 h at 37 °C. The fluorescent images were obtained using a LSM980 confocal fluorescence microscope (Zeiss, Oberkochen, Germany) to observe the uptake of exosomes by BMSCs.

Cell Counting Kit-8 Assay

The cell counting Kit-8 (CCK8) reagent (Dojindo, Kumamoto, Japan) was used for this assay. Briefly, mouse BMSCs were seeded in 96-well plates at 3 000 cells/100 μL/well. After allowing 24 h for cell apposition, the medium was changed to a serum-free medium containing different concentrations of exosomes at 0, 25, 50, and 100 μg/mL. The cells were incubated for 24 and 48 h, respectively, then 100 μL of working solution (CCK8 solution:serum-free medium = 1:9) was added then the 96-well plates were incubated for 2 h at 37°C. The absorbance at 450 nm was measured using a microplate reader (Thermo, USA).

Migration Assay

A culture-insert in µ-Dish 35 mm (Ibidi, Martin Reid, Germany) was placed inside the 24-well plate. The BMSCs were digested and adjusted to a density of 6 × 10⁶ cells/mL. Cell suspension with 100 μL was added to each well of the culture-insert, and cell suspension with 300 μL was added outside of the culture-insert, avoiding moving the culture-insert during the process of adding liquid. The culture-insert was removed after 24 h of incubation at 37 °C. Then, the unattached cells were carefully washed off using PBS 3 times. After changing to serum-free medium containing 100 μg/mL exosomes, images were recorded using an ordinary light microscope at 0, 6, 12, and 24 h, respectively, after scratching. Final image processing was performed using ImageJ (National Institutes of Health, USA).

Alizarin Red Staining

Briefly, mouse BMSCs were seeded in 48-well plates at 5 000 cells/200 μL/well. Exosomes were added to the medium to reach a concentration of 100 μg/mL. On day 14th, cells were carefully washed twice using PBS and then 200 μL of 4% paraformaldehyde was added to each well for 30 min. Then alizarin red staining solution (Solarbio) was then added to each well and incubated for 30 min. Finally, the residual staining solution was removed by washing with ultrapure water for 5 min. After drying at room temperature, the red mineralized nodules were observed under a stereomicroscope and photographed.

ALP Staining and Activity

Mouse BMSCs were seeded in 48-well plates at 5 000 cells/200 μL/well. The medium containing 100 μg/mL exosomes was changed when the cell density reached 70%-80%. The medium was changed every 2 days. On the 7th day, the medium was discarded and the cells were carefully washed twice using PBS. Each well was fixed with 200 μL of 4% paraformaldehyde for 30 min. After 24 h of color development using BCIP/NBT ALP working solution (Beyotime, China), the color development reaction was terminated by washing 3 times with ultrapure water, and the cells were dried at room temperature. The images were recorded using a stereomicroscope (Leica, Germany).

The cells were lysed using 1% TritonX-100 on ice for 30 min, then the cell lysate was collected and the supernatant was obtained after centrifugation at 10 000g for 5 min. The absorbance of each well was measured at 520 nm using an enzyme digester after treatment according to the ALP activity test kit (Nanjing Jiancheng Biology) operation table. The protein concentration of the lysed cells was quantified using the BCA protein assay kit. Finally, ALP activity (Kim’s U/gprot) was obtained according to a formula from the ALP activity test kit specification.

Western Blot Analysis

After culture for 7 days, the supernatant from BMSCs in 100 μg/mL exosome medium was taken after lysing the cells using RIPA lysis buffer (Beyotime). The supernatant was denatured to the protein after 10 min in a 95°C water bath. Protein concentration was quantified using the BCA protein assay kit. The target proteins were transferred to the polyvinylidene fluoride (PVDF) membrane by electrophoresis on SDS-PAGE gels. After being blocked for 15 min using a blocking buffer (Beyotime, China), the PVDF membrane was incubated overnight at 4°C using primary antibodies anti-Runx2 (1:1000; Cell Signaling Technology, USA), anti-BMP2, anti-ALP, and anti-GAPDH (1:1000; Abcam, England). The secondary antibodies were then incubated on a shaker at room temperature for 2 h. Finally, protein expression levels were detected using an Odyssey CLx imaging system (LI-COR Biosciences), and protein band density was semi-quantified using ImageJ.

Verification of the BMSC Aptamer Affinity

By reviewing the literature, a BMSC aptamer sequence was obtained.³⁵ To verify the high specificity of this aptamer for BMSCs, digested BMSCs and bone marrow-derived cells (BMDCs) of C57BL/6J mice were adjusted to a concentration of 10 × 10⁷/mL. 100 μL of cell suspension was incubated with 200 nM of Alexa 647 fluorescently labeled BMSC aptamers and scrambles in serum-free culture medium for 3 h at 37 °C. After incubation, the cells were washed twice with PBS. Finally, the cells were suspended using 300 μL PBS, passed through a 400 MeSH filter and fluorescent signals were detected using a CytoFLEX flow cytometry (Beckman Coulter). The binding rate of BMSC aptamers to target cells was assayed using no treated cells as a reference and scrambles as a negative control. FlowJo software (version: FlowJo v10, Leonard Herzenberg, USA) was used for data analysis.

The specificity of this aptamer with BMSCs was further confirmed using confocal fluorescence microscopy as well. After 2 × 10⁵ cells were spread in 15 mm confocal dishes overnight, the original medium was removed and replaced with Alexa 647 fluorescently labeled BMSC aptamers or scrambles at a concentration of 400 nM in serum-free medium for 4 h. After incubation, the cells were washed twice with PBS for 5 min each time and fixed with 4% paraformaldehyde for 30 min. After fixation, the cytoskeletons were stained with Actin-Tracker Green (Beyotime) for 30 min and the nuclei were stained with 4ʹ,6-diamidino-2-phenylindole (DAPI; Beyotime) for 15 min. Finally, the fluorescent images were obtained using a LSM980 confocal fluorescence microscope.

The sequence of the BMSC aptamer was as follows:

5ʹ-ACGACGGTGATATGTCAAGGTCGTATGCACGAGTCAGAGG-3ʹ

The sequence of the BMSC scramble was as follows:

5ʹ-GAGTATATGTTAGGCCTGGGTGAGTCCTTGCGTCTTCTA-3ʹ

Construction and Synthesis of 3WJ RNA Nanoparticles

The 3WJ RNA nanoparticles were synthesized by bottom-up self-assembly. The 3WJ-BMSC_apt-Chol RNA nanoparticles consisted of 3 chains, a_3WJ-Cholesterol, b_3WJ-BMSC aptamer, and c_3WJ-Alexa 647, as shown in the schematic diagram (Fig. 3E). Chain a connected to cholesterol was used to connect to the exosome membrane attachment, chain b attached BMSC aptamers as targeting ligand, and chain c attached to Alexa 647 as an imaging module. The chain b without BMSC aptamers attached was used as a negative control (3WJ-Chol) (Fig. 3G). Each strand was mixed 1:1:1 by molarity and then dropped from 95 to 4°C at 0.1°C per second using a PCR instrument. The prepared 3WJ RNA nanoparticles were tested for assembly efficiency by natural polyacrylamide gel electrophoresis (PAGE). The sequence of each strand is shown subsequently (lowercase letters indicate 2ʹOMe modified bases).

Figure 3.

Specific binding of aptamers to BMSCs and 3WJ RNA nanoparticles harboring BMSC aptamer and physicochemical characteristics of 3WJ-BMSC_apt/M2-Exos. (A) Flow cytometry analysis of the specific Alexa 647-labeled BMSC aptamer on BMSCs. (B) Flow cytometry analysis of the specific Alexa 647-labeled BMSC aptamer on BMDCs. (C) Representative confocal microscopic image of the Alexa 647-labeled BMSC aptamer cocultured with BMSCs, scale bar: 50μm. Alexa 647: 3WJ RNA nanoparticles; Actin: cytoskeletons; DAPI: the cell nucleus. (D and E) PAGE fluorescence image and schematic illustration of 3WJ-BMSC_apt-Chol RNA nanoparticles. (F and G) PAGE fluorescence image and schematic illustration of 3WJ-Chol RNA nanoparticles. (H) Fluorescence images of M2-Exos, Alexa 647-3WJ/M2-Exos, and Alexa 647-3WJ-BMSC_apt/M2-Exos, scale bar: 5 μm. Alexa 647: red, 3WJ RNA nanoparticles; DiO: green, exosomes. (I) Transmission electron micrograph of M2-Exos, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos, scale bar: 100 nm. (J) Size distributions of M2-Exos, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos by NTA. (K) WB analysis of the characteristic exosomal proteins including Alix, Flot, and HSP70 from M2-Exos, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos.

a_3WJ-Cholesterol: 5ʹ-cGG uAG cAc GGG cuG uGc G (Cholesterol TEG)-3ʹ

b_3WJ: 5ʹ-cGc AcG Acc GAc AcG c-3ʹ

b_3WJ-BMSC aptamer: 5ʹ-cGc AcG Acc GAc AcG cAC GAC GGT GAT ATG TCA AGG TCG TAT GCA CGA GTC AGA GG-3ʹ (the underlined sequence is BMSC aptamer).

c_3WJ-Alexa 647: 5ʹ-GcG uGc uGG uGc uAc cG-Alexa 647-3ʹ

Conjugation of 3WJ RNA Nanoparticles to Exosomes and Characterization

0.56 nmol 3WJ-BMSC_apt-Chol RNA nanoparticles or 3WJ-Chol RNA nanoparticles were added to 100 μg exosomes and incubated for 45 min at 37 °C, then for 1 h on ice. Unconjugated 3WJ RNA nanoparticles were removed by ultracentrifugation at 120 000g for 70 min and the 3WJ RNA nanoparticles conjugated exosomes were harvested (3WJ-BMSC_apt/M2-Exos or 3WJ/M2-Exos). The modified exosomes were stored at −80°C until use. To assess whether exosomes are successfully conjugated with nanoparticles, Alexa 647 conjugated exosomes were stained by DiO, and exosomes that did not react with 3WJ RNA nanoparticles were used as controls. The fluorescent images were obtained using a fluorescence microscope (Zeiss, Oberkochen, Germany). TEM was employed to observe the morphology of 3WJ-BMSC_apt/M2-Exos, 3WJ/M2-Exos, and M2-Exos. NTA was applied to analyze the diameter and size distribution of 3WJ-BMSC_apt/M2-Exos, 3WJ/M2-Exos, and M2-Exos. The protein expression levels of exosomal signature proteins of 3WJ-BMSC_apt/M2-Exos, 3WJ/M2-Exos, and M2-Exos were detected using WB analysis.

In vitro Targeting Abilities of Exosomes

To examine the targeting abilities of the modified exosomes in vitro, BMSCs were seeded in 6-well plates at 2 × 10⁵ cells/well. When the cell density reached 80%-90%, 50 μg/mL Alexa 647 fluorescently labeled 3WJ-BMSC_apt/M2-Exos or 3WJ/M2-Exos were incubated with BMSCs at 37°C for 3 h. After that, the cells were washed twice times with PBS. Finally, the cells were re-suspended using 300 μL PBS and passed through a 400 MeSH filter. The fluorescent signals in cells were acquired using a CytoFLEX flow cytometry. FlowJo software was used for data analysis.

To verify the uptake of the modified exosomes by BMSCs, 2 × 10⁵ cells were seeded in 15 mm confocal dishes and cultured overnight. 50 μg/mL Alexa 647 fluorescently labeled 3WJ-BMSC_apt/M2-Exos or 3WJ/M2-Exos were incubated with BMSCs at 37°C for 4 h. After incubation, the cells were washed twice (5 min/time) with PBS and fixed in 4% paraformaldehyde for 30 min. Subsequently, the cytoskeletons were stained with Actin-Tracker Green for 30 min and the nuclei were stained with DAPI for 15 min. Finally, the fluorescent images were obtained using a LSM980 confocal fluorescence microscope.

In vivo Imaging of Biodistribution

Alexa 647 fluorescently labeled 3WJ RNA nanoparticle-modified exosomes were used to test the distribution by the in vivo imaging system. The mice were depilated on all extremities before administration. 100 μg/200 μL 3WJ-BMSC_apt/M2-Exos or 3WJ/M2-Exos were injected into mice via the tail veins in a single injection. The targeting of 3WJ-BMSC_apt/M2-Exos was clarified by detecting the changes of fluorescence intensity in limbs at 2, 6, 12, and 24 h after administration by an IVIS Spectrum imaging system (PerkinElmer, USA). To observe the distribution of the modified exosomes in the major organs, the mice were anesthetized and sacrificed 12 h-post exosome injection. The distribution of near-infrared fluorescence signals in the heart, lung, liver, spleen, kidney, and bone of mice in each group was detected using the Odyssey CLx imaging system.

The Treatment of Mice Femoral Fracture

All animal experiments were approved by the Ethics Committee of Guangzhou Medical University (GY2021-021). C57BL/6J mice (female, 12-week-old) were selected as our experimental animals. The mice were fasted from food and water for 18-24 h before surgery. The mice were anesthetized by intramuscular injection with Zoletil50 (80-100 μg/animal). After that, the left limb of the mice and the operative areas were shaved and disinfected with complex iodine (Fig. 5A①). A transverse incision was made along the skin of the knee joint with ophthalmic scissors, and the knee joint was flexed to fully expose the white tendon of the knee joint (Fig. 5A②). The femur was exposed by blunt separation along the parallel direction of the femur with ophthalmic forceps (Fig. 5A③). Next, the femur was held with forceps, and a 22-gauge syringe needle was inserted vertically into the intercondylar fossa of the femur into the femoral marrow (Fig. 5A④). A transverse fracture was formed in the middle femur using bone forceps (Fig. 5A⑤). The excess needle was cut off with ophthalmic forceps (Fig. 5A⑥). Finally, the wound was closed layer by layer, and the muscle was closed with absorbable sutures, while the skin was sutured with 4-0 nylon thread (Fig. 5A⑦,⑧). After the surgery was completed, 10 μL of sulforaphane was injected intramuscularly. Cephalosporin was injected intramuscularly daily for 3 days post-operation to prevent infection. The mice were randomly divided into 3 groups: PBS (n = 6), 3WJ/M2-Exos (n = 6), and 3WJ-BMSC_apt/M2-Exos (n = 6). Three days after surgery, 100 μL 3WJ/M2-Exos or 3WJ-BMSC_apt/M2-Exos or an equivalent volume of PBS was injected every 3 days through the tail vein. Mice were sacrificed on the 15th day after treatment. The femurs were then harvested and fixed in 4% paraformaldehyde for subsequent experiments.

Figure 5.

Administration 3WJ-BMSC_apt/M2-Exos facilitate femur bone fracture healing in a mouse model. (A) Surgical photographs of the mouse femur fracture model. (B) Representative radiographs of the mouse femur fracture model at 24 h, 3 days, 7 days, and 14 days. Arrows indicate the fracture site. (C) Representative 3D reconstruction images, 2D cross-sectional and sagittal images from micro-CT scanning of the mouse femur fracture model. (D) Quantitative analysis of BV/TV, BMD, Tb. N and Tb. Sp via micro-CT (n = 4). Error bar, mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001.

X-ray Radiography

X-ray radiography of the left femur of mice was obtained by the IVIS Spectrum imaging system X-ray module at 24 h, 3, 7, and 14 days after surgery. The exposure time of the shots was 20 s. The images were captured and analyzed by the imaging unit.

Micro-CT Analysis

A ZKKS-MCT-Sharp Micro-CT scanner (Guangzhou, China) was used to assess the healing of femur fracture in mice. For scanning, a sample was fixed in the fixator along the long axis, with a 70 KV scanning voltage, 7 W power, 4-frame superposition, 0.72° angular gain, and 100 ms exposure time. The sample was rotated for one week to complete the scan. The fracture region was selected as the region of interest, and the 3D image was obtained. Measured parameters included the bone volume/total volume (BV/TV), bone mineral density (BMD), trabecular number (Tb. N), and trabecular separation (Tb. Sp). Micro-CT results were analyzed using a ZKKS-Micro CT 4.1 software.

Histological and Immunohistochemical Examination

Histological and immunohistochemical examinations were performed as described previously. Briefly, after fixation with 4% paraformaldehyde for 2 days, the samples were decalcified in 10% ethylenediaminetetraacetic acid (EDTA) at room temperature for 4 weeks. Next, the sample tissues were embedded in paraffin after step-by-step dehydration through different gradient concentrations of alcohol. Paraffin sections of approximately 4 μm in thickness were obtained using a Leica RM2016 slicer. Subsequently, the sections were subjected to H&E staining and Masson trichrome staining to identify the formation of new bone. To analyze the osteogenic activities of the samples, paraffin sections were dewaxed first and hydrated with xylene and gradient alcohol, and then tissue sections were placed in a repair cassette filled with citric acid antigen repair buffer (pH 6.0) in a microwave oven for antigen repair. After closure using goat-derived rabbit serum, tissue sections were incubated with primary antibody ALP and osteocalcin (OCN; Sigma) at 4°C overnight. Sections were washed 3 times with PBS followed by the addition of secondary antibodies and incubated for 50 min at room temperature. Finally, the color was developed using a diaminobenzidine (DAB) color development solution (Solarbio, Beijing, China) and hematoxylin re-stained cell nuclei. Images were obtained by a digital pathology slide scanner (Leica CS2, Germany). Image processing was performed using ImageScope (Leica, Germany).

Small RNA Sequencing Analysis

Exosomal RNA isolation was extracted using the exoRNeasy Serum/Plasma Maxi Kit (QIAGEN, Germany). The libraries were constructed using QIAseq miRNA Library Kit (QIAGEN, Germany). Libraries were multiplexed and sequenced on an Illumina NextSeq CN500 platform (Illumina, USA). Fastq files were generated after sequencing. The raw data were quality pre-processed using Trimmomatic software to remove adaptor sequences and low-quality bases and miRNAs were identified with the miRBase database. The gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the top 50 miRNAs were performed using clusterProfiler_3.6.0.

Statistical Analysis

All data were analyzed by GraphPad Prism (version 8.0; GraphPad Software, USA), and values were expressed as mean ± SD. Differences between 2 groups were analyzed using t test while the differences among multiple groups were analyzed using 1-way analysis of variance (ANOVA). P < .05 indicated statistically significant.

Results

Characterization of M2-Exos

After 24 h, macrophages in the IL-4-treated group had more and longer tentacles than the PBS-treated group though cells in both groups had similar general morphology (Fig. 1A). Flow cytometry analysis showed that the double-positive rates of F4/80 and CD206 were significantly higher in IL-4-treated macrophages than in PBS-treated macrophages (Fig. 1B). These results demonstrated the successful polarization of M2 macrophages.

Figure 1.

Characterization of M2 macrophages and M2-Exos. (A) Morphology of RAW264.7 (M0 macrophages) and IL-4 polarized M2 macrophages, scale bar: 500 μm. (B) Flow cytometry analysis of M2 macrophage markers F4/80 and CD206. (C) Transmission electron micrograph of M2-Exos, scale bar: 100 nm. (D) Size distributions of M2-Exos by NTA. (E) WB analysis of the characteristic exosomal proteins including Alix, Flot, and HSP70 from M2-Exos. (F) Confocal microscopic image of DiO-labeled exosomes cocultured with BMSCs, scale bar: 50 μm. DAPI: cell nucleus; DiO: exosomes.

Subsequently, we extracted M2-Exos using ultracentrifugation and characterized the collected M2-Exos by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and western blot (WB). TEM showed that M2-Exos had a typical cup-shaped morphology with a bilayer membrane structure (Fig. 1C). NTA revealed that the diameter distribution of M2-Exos was in the range of 40-160 nm, which was consistent with the diameter range of exosomes (Fig. 1D).¹⁷ WB showed that the exosomes were enriched in exosomal characteristic proteins, such as Alix, Flot, and HSP70, while endoplasmic reticulum specific protein-Calreticulin was not detected in exosomes (Fig. 1E). To examine whether M2-Exos could be internalized by BMSCs, the membrane fluorescent dye DiO-labeled exosomes were incubated with BMSCs for 4 h. Then a large amount of green fluorescence could be seen within BMSCs (Fig. 1F), suggesting that M2-Exos have the ability to be taken up and internalized by BMSCs.

M2-Exos Promoted the Proliferation, Migration, and Osteogenic Differentiation of BMSCs In vitro

Using the CCK8 test, we found that the OD value of the 100 μg/mL M2-Exos group was the largest at either 24 or 48 h, and increased with time (Fig. 2A). This indicates that 100 μg/mL M2-Exos significantly promoted the proliferation of BMSCs, and thus this concentration was used for subsequent experiments. The migration assay (Fig. 2B and C) was used to demonstrate whether M2-Exos could promote the migration of BMSCs. As shown in Fig. 2B and C, the healing area was greatly increased in the M2-Exos group at 12 and 24 h compared with the control group without the addition of exosomes, and the cell crawling in the M2-Exos group almost completely occupied the scratch area at 24 h. Significant differences in the healing area between the M2-Exos group and the control group were detected at 12 and 24 h. However, there was no statistical difference in the migration rate between the 2 groups at 6 h.

Figure 2.

M2-Exos enhanced the proliferation, migratory and osteogenic activity of BMSCs. (A) CCK8 results of M2-Exos after 24 and 48 h, respectively, incubation with BMSCs using different concentrations. (B and C) Migration assay and quantitative analysis of the M2-Exos group at 0, 6, 12, and 24 h, scale bar: 500 μm. (D) Alizarin red staining mineralization of M2-Exos after 14 days of coculture with BMSCs, scale bar:400 μm. (E) ALP staining of M2-Exos after 7 days of coculture with BMSCs, scale bar:400 μm. (F) ALP activity of M2-Exos after 7 days of coculture with BMSCs. (G and H) WB and quantitative analysis of osteogenic proteins including ALP, BMP2, and Runx2 from M2-Exos. Error bar, mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001; ns, not significant.

Alizarin red staining was performed after culturing BMSCs under different media for 14 days. More mineralized nodules were observed in the M2-Exos group compared to the PBS control group (Fig. 2D). Similarly, ALP staining and ALP activity were found to be stronger in the M2-Exos group than in the PBS control group (Fig. 2E and F). In addition, the expression of M2-Exos osteogenic proteins ALP, BMP2, and Runx2 were higher than that of the PBS control group by WB analysis (Fig. 2G and H). These findings suggest that M2-Exos were able to stimulate the proliferation and migratory ability of BMSCs, as well as to promote osteogenic differentiation of BMSCs in vitro.

Construction of 3WJ RNA Nanoparticles Harboring BMSC Aptamer and Characterization of 3WJ-BMSC_apt/M2-Exos

Next, we employed flow cytometry analysis to assess the binding rate using BMSCs and BMDCs with Alexa 647-labeled aptamers and scrambles. Specifically, the positive fluorescence signals in Alexa 647-labeled BMSC aptamer-treated BMSCs were stronger than in Alexa 647-labeled BMSC aptamer-treated BMDCs (Fig. 3A and B). Confocal fluorescence microscope showed that a large amount of red fluorescent signal was visible in BMSCs co-incubated with BMSC aptamers, not BMSC scrambles (Fig. 3C). To modify this BMSC aptamer to M2-Exos simply and rapidly, we designed 2 3WJ RNA nanoparticles, 3WJ-BMSC_apt-Chol RNA nanoparticles (Fig. 3E), and negative control 3WJ-Chol RNA nanoparticles (Fig. 3G), respectively. Both 3WJ RNA nanoparticles were fluorescently photographed by PAGE electrophoresis against the gel map and no byproduct bands were found (Fig. 3D and F), suggesting that the synthesized 3WJ RNA nanoparticles were well assembled.

To verify whether M2-Exos was successfully conjugated to 3WJ RNA nanoparticles, the membrane fluorescent green dye DiO was used to label M2-Exos, Alexa 647-3WJ/M2-Exos, and Alexa 647-3WJ-BMSC_apt/M2-Exos, only green fluorescence was found to be present under fluorescence microscopy for M2-Exos, while Alexa 647 and DiO co-localization was found in Alexa 647-3WJ/M2-Exos and Alexa 647-3WJ-BMSC_apt/M2-Exos (Fig. 3H). This demonstrated that we successfully modified M2-Exos. Subsequent morphology of M2-Exos, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos were examined by TEM. Specifically, all 3 exosomes had a typical cup-shaped morphology with intact membranes and no damage, but the diameter of the modified exosomes was slightly larger than that of the unmodified exosomes (Fig. 3I). NTA also found a similar distribution of unmodified and modified exosome diameters with peaks of 108 nm, 115 nm, and 121 nm, respectively (Fig. 3J). According to the WB analysis, both unmodified and modified exosomes were enriched in exosomal characteristic proteins Alix, Flot, and HSP70, and the expression was similar (Fig. 3K). These results indicate that the modified exosomes still presented intact vesicle shape, and were as rich in exosomal characteristic proteins as unmodified exosomes.

Targeting of 3WJ-BMSC_apt/M2-Exos to BMSCs In vitro and to the Fracture Site In vivo

To validate 3WJ-BMSC_apt/M2-Exos targeting to BMSCs, in vitro experiments, flow cytometry analysis was employed to assess the binding rate of 3WJ-BMSC_apt/M2-Exos to BMSCs. We found that the positive fluorescence signals in Alexa 647-labeled 3WJ-BMSC_apt/M2-Exos-treated BMSCs were stronger than in Alexa 647-labeled 3WJ/M2-Exos-treated BMSCs (Fig. 4A). According to confocal fluorescence microscope analysis, 3WJ-BMSC_apt/M2-Exos showed apparent uptake to BMSCs compared to 3WJ/M2-Exos (Fig. 4B).

Figure 4.

Targeted delivery of 3WJ-BMSC_apt/M2-Exos into BMSCs and the fracture site. (A) Flow cytometry analysis of Alexa 647-labeled 3WJ/M2-Exos and 3WJ-BMSC_apt/M2-Exos uptake by BMSCs. (B) Representative confocal microscopic image of cellular uptake of Alexa 647-labeled 3WJ/M2-Exos or 3WJ-BMSC_apt/M2-Exos cocultured with BMSCs, scale bar: 50 μm. Alexa 647: 3WJ/M2-Exos or 3WJ-BMSC_apt/M2-Exos; Actin: cytoskeletons; DAPI: the cell nucleus. (C and D) Fluorescence distribution and intensity changes of Alexa 647-labeled 3WJ/M2-Exos and 3WJ-BMSC_apt/M2-Exos in mice at different periods (n = 3). (E) Representative distribution images of 3WJ/M2-Exos and 3WJ-BMSC_apt/M2-Exos in major organs and limbs (n = 3). (F) Relative Alexa 647 intensity in each organ. Error bar, mean ± SD. *P < .05; **P < .01; ns, not significant.

Similarly, the in vivo biodistribution of 3WJ-BMSC_apt/M2-Exos is important for subsequent therapeutic efficacy. To assess the ability of the modified exosomes in targeting the limbs in vivo, we injected 100 μg of Alexa 647-labeled 3WJ-BMSC_apt/M2-Exos and 3WJ/M2-Exos in a tail vein injection in C57BL/6J mice and observed the fluorescence distribution of the injected exosomes in vivo in mice using an IVIS Spectrum imaging system. As expected, we observed that the exosomes were mainly distributed in the limbs after injection, and the fluorescence intensity was stronger in mice injected with 3WJ-BMSC_apt/M2-Exos than in those injected with 3WJ/M2-Exos (Fig. 4C and D). To explore the fluorescence intensity distribution of modified exosomes in the major organs, mice were dissected 12 h after exosome injection. Then, the major organs, spine, and limbs were removed, and it was found that the fluorescence of mice injected with 3WJ-BMSC_apt/M2-Exos was mainly concentrated in the limbs, and the fluorescence intensity was significantly stronger than that of mice injected with 3WJ/M2-Exos (Fig. 4E and F). Overall, 3WJ-BMSC_apt/M2-Exos can target BMSCs in vitro and the fracture site in vivo.

3WJ-BMSC_apt/M2-Exos Accelerate Bone Fracture Healing in Mice

A mouse femur fracture model was used to investigate 3WJ-BMSC_apt/M2-Exos for accelerated bone fracture healing. The fracture healing ability of the mice changed greatly in the second and third weeks, and obvious bone callus could be formed (Supplementary Fig. S1). A 2-week window was chosen to observe the fracture healing. We monitored bone fracture healing using X-ray and Micro-CT analysis. Specifically, significant bone callus formation was obvious in all 3 groups, 3WJ-BMSC_apt/M2-Exos, 3WJ/M2-Exos, and PBS groups, on days 14 post-surgery, but the fracture gap at the femur fracture became smaller on the 3WJ-BMSC_apt/M2-Exos group compared with the other 2 treatment groups (Fig. 5B). Furthermore, representative 3D reconstruction images of micro-CT are consistent with the X-ray findings. According to the micro-CT 2D cross-sectional and sagittal images at the fracture site, we found that the cortical bone at both ends of the fracture site was discontinuous and the surrounding bone callus was more extensive cancellous bone in the PBS group, followed by the 3WJ/M2-Exos group, and denser bone callus in the 3WJ-BMSC_apt/M2-Exos group (Fig. 5C). When the bone callus composition at the fracture site was quantified, the 3WJ-BMSC_apt/M2-Exos group exhibited higher BV/TV, BMD, and Tb. N, but Tb. Sp was the lowest compared with the other 2 treatment groups (Fig. 5D).

In addition, post-decalcification samples were subjected to histological analysis, including HE and Masson trichrome staining, to identify new bone formation at the fracture site, and it was found that 3WJ-BMSC_apt/M2-Exos had more new bone formation at the bone callus than the other 2 groups (Fig. 6A). In addition, abundant bone trabeculae were observed in the 3WJ-BMSC_apt/M2-Exos group. Moreover, there was plentiful fibrous connective tissue in the other 2 groups. ALP and OCN were 2 representative osteogenic markers. Immunohistochemical staining revealed that both ALP and OCN were more positively produced in the 3WJ-BMSC_apt/M2-Exos group than in the other groups (Fig. 6B). The quantitative analysis of new bone, ALP and OCN showed that there were significant differences among all groups (Fig. 6C).

Figure 6.

Histological and immunohistochemical analysis of ALP and OCN. (A) Representative images of HE and Masson trichrome staining at the femoral fracture site in each group (scale bar: 1 mm for low magnification, and 100 μm for enlargement). (B) Representative images of ALP and OCN immunohistological staining at the femoral fracture site in each group (scale bar: 1 mm for low magnification, and 100 μm for enlargement). (C) Quantitative analysis of the positive production area of ALP and OCN in newly formed bone and the new bone formation among all groups (n = 4). Error bar, mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001.

The miRNA Expression Profile Analysis in 3WJ-BMSC_apt/M2-Exos

To identify the osteogenesis mechanism, small RNA sequencing was conducted in M2-Exos and 3WJ-BMSC_apt/M2-Exos to resolve the detailed components carried by these exosomes. The preliminary results of heatmap analysis showed that there was little difference in miRNA expression between M2-Exos and 3WJ-BMSC_apt/M2-Exos among the top 50 significantly expressed miRNAs (Fig. 7A). We have found that miR-221-3p was significantly decreased in 3WJ-BMSC_apt/M2-Exos. The downregulation of miR-221-3p can promote the anti-inflammatory direction of M2 macrophages, which is conducive to bone formation.³⁷ Furthermore, let-7i-5p,³⁸ miR-29a-3p,³⁹ miR-25-3p,⁴⁰ miR-690,¹⁵ miR-378a-3p²⁰ and miR-140-3p,⁴¹ which are enhancers of osteogenic differentiation, were highly expressed in both M2-Exos and 3WJ-BMSC_apt/M2-Exos. GO was used to study the biological processes (BP), cellular components (CC), and molecular functions (MF) regulated by the top 50 miRNAs in the exosomes (Fig. 7B-D). GO analysis indicated that the miRNAs were associated with “positive regulation of transcription from RNA polymerase II promoter” and “transcription from RNA polymerase II promoter” in BP, “cytoplasm” and “membrane” in CC, and “protein binding” and “sequence-specific DNA binding” in MF. Subsequent KEGG pathway analysis showed that the miRNAs were involved in important signaling pathways for osteogenic differentiation, such as the MAPK signaling pathway,⁴² Wnt signaling pathway,⁴³and PI3K-Akt signaling pathway⁴⁴ (Fig. 7E).

Figure 7.

miRNAs analysis in M2-Exos and 3WJ-BMSC_apt/M2-Exos. (A) A heatmap shows the top 50 significantly expressed miRNAs of M2-Exos and 3WJ-BMSC_apt/M2-Exos. (B-D) GO analysis of BP, CC, and MF of the top 50 significantly expressed miRNAs. (E) KEGG pathway analysis of the top 50 significantly expressed miRNAs. GO: gene ontology; BP: biological processes; CC: cellular components; MF: molecular functions; KEGG: Kyoto Encyclopedia of Genes and Genomes.

Discussion

Macrophages are innate immune cells that play an important role in maintaining cellular homeostasis and promoting bone tissue regeneration. Particularly, M2 macrophages cooperate with BMSCs in the hierarchical processes of bone fracture healing.⁴⁵ Exosomes are extracellular vesicles that are involved in intercellular communication mainly by delivering bioactive molecules. Recently, the regulation effect of M2-Exos on tumors, diabetes, cardiac infarction, and atherosclerosis has been confirmed.⁴⁶ In addition, the role of M2-Exos in bone homeostasis and remodeling has been gradually recognized.⁴⁷ Our study showed that M2-Exos could be internalized by BMSCs and stimulate the proliferation and migration ability of BMSCs. Besides, M2-Exos also promotes bone mineral deposition and osteogenic differentiation of BMSCs in vitro. These results suggest that M2-Exos are an important mediator to accelerate the osteogenic differentiation of BMSCs.

The effect of exosomes in vivo is closely related to the origin of exosomes and the method of administration.²³ Previous studies have shown that the injection of exosomes in situ can promote bone fracture healing.^48,49 However, due to rapid metabolism and clearance, local administration of exosomes due to rapid metabolism and clearance may result in limited clinical efficacy. Besides, patients with delayed union or nonunion are expected to receive systematic administration to enhance or accelerate bone fracture healing. Systematic administration is a noninvasive manner while avoiding reoperation and allowing for continuous drug delivery and long-term stimulation of bone fracture healing.^50,51 However, the poor targeting ability of unmodified exosomes restricts their usage in clinics. Many studies have emphasized that functional modification can enhance the targeting ability of exosomes thereby increasing their local concentration at the lesion site, reducing their toxicity and side effects, and maximizing therapeutic efficacy.^52-55 Hence, we tried to explore an effective way of exosome modification in the treatment of bone fracture through intravenous injection.

We presented a simple and rapid surface modification of exosomes using 3WJ RNA nanoparticles. Because exosomes have lipid bilayers with high cholesterol content, cholesterol has a strong affinity for exosomes.⁵⁶ Based on this property, the multifunctional 3WJ RNA nanoparticles containing cholesterol structure only use 3 h to conjugate M2-Exos and do not require any catalyst, thus ensuring their safety. This is advantageous over other traditional exosome chemical modification methods, such as click chemistry which require more than 12 h to synthesize engineered exosomes.^52,54 With the aid of 3WJ nanostructure, RNA nanotechnology can conjugate aptamers for specific cells and therapeutic modules by bottom-up self-assembly, which has been widely used in drug delivery, tumor treatment, etc.^27,57 Moreover, the size, shape, and stoichiometry of multifunctional 3WJ RNA nanoparticles can be precisely regulated. Their negatively charged and hydrophilic nature make them avoid accumulation in the liver, thus reducing potential toxicity and immunogenicity.^58,59 In this study, we first used RNA nanotechnology to prepare multifunctional 3WJ RNA nanoparticles for bone fracture healing. Our results indicate that the BMSC aptamer is highly specific for BMSCs, and 3WJ RNA nanoparticles with BMSC aptamer strands were prepared greatly for subsequent modification and targeting of M2-Exos. The 3WJ RNA nanoparticles contain a cholesterol structural domain bound to exosome membranes, targeting modules with BMSCs aptamers, and imaging modules with Alexa 647. The multifunctional 3WJ RNA nanoparticles were subsequently conjugated to M2-Exos as a therapeutic module, and the final aptamer-functionalized exosomes 3WJ-BMSC_apt/M2-Exos (or 3WJ/M2-Exos without BMSCs aptamers as controls) were obtained. We exploited the unique osteogenic bioactivity of M2-Exos to regulate BMSCs property, taking advantage of the combination of RNA nanotechnology and immune cell-based intrinsic function to mediate osteogenesis. The method is suitable for modifying exosomes of any cellular origin. Furthermore, multiple targeted ligands and therapeutic modules can be simultaneously combined for better targeting and therapy.

A potential concern when using RNA nanotechnology to modify exosomes is that this may alter the original characteristics of exosomes, affecting the biological function of exosomes. In this study, the morphology, size, and exosomal characteristic proteins of M2-Exos, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos were characterized and measured, respectively. Our results showed that although the diameter size of the modified exosomes increased slightly, the exosomal membrane remained intact and the exosomal characteristic proteins of modified exosomes were as enriched as the unmodified exosomes. This also suggests that aptamer-functionalized exosomes have the potential for in vivo application, which might improve the therapeutic efficacy and facilitate early bone fracture healing.

After the successful synthesis of 3WJ-BMSC_apt/M2-Exos with maintained exosome characteristics, we co-incubated Alexa 647-labeled 3WJ-BMSC_apt/M2-Exos with BMSCs. Our data indicate that the positive fluorescence signals and the percentage of Alexa 647-positive cells were stronger in 3WJ-BMSC_apt/M2-Exos-treated BMSCs. These findings are consistent with previous studies,³⁵ and suggest the targeting ability of 3WJ-BMSC_apt/M2-Exos was mediated by the specific binding of the aptamer to BMSCs. Furthermore, in vivo biodistribution after 3WJ-BMSC_apt/M2-Exos injection by tail vein revealed that 3WJ-BMSC_apt/M2-Exos could be more specifically targeted to the limbs, with higher concentrations and little or no accumulation into healthy organs and tissues compared to the control treatments without BMSC aptamers. Also, compared to the shorter half-life of bare exosomes,⁶⁰ 3WJ-BMSC_apt/M2-Exos stayed longer in the body after injection in the fracture site. These results suggest that 3WJ-BMSC_apt/M2-Exos were well targeted to the fracture site in vivo. Next, bone fracture healing was determined based on the femoral fracture gap in mice, the size of the bone callus formation around the fracture gap, and quantitative analysis of the bone callus composition after systemic administration of PBS, 3WJ/M2-Exos, and 3WJ-BMSC_apt/M2-Exos. Bone formation after femur fracture in mice was further confirmed by histological analysis. The results of our data indicated that the 3WJ-BMSC_apt/M2-Exos group had a more pronounced effect on promoting femur bone fracture healing compared to the other 2 groups. Overall, we have successfully targeted aptamer-functionalized exosomes as a drug delivery system to the fracture site and accelerated femur bone fracture healing in mice. Based on this 3WJ RNA nanoparticle modification and the inherent immune role of macrophages, M2-Exos can also be modified with many other components such as siRNA, other aptamers, and chemotherapeutic agents, which can be used for targeted therapy of additional diseases such as osteoporosis, rheumatoid arthritis, myocardial infarction, and cancer.

Exosomes transmit paracrine signals through miRNAs during bone regeneration.²⁰ According to the miRNA expression profile analysis in 3WJ-BMSC_apt/M2-Exos, miR-29a-3p, miR-25-3p and miR-690, which are responsible for osteogenic differentiation, were riched in 3WJ-BMSC_apt/M2-Exos. Moreover, recent studies have shown that miR-29a-3p and miR-25-3p are associated with the Wnt signaling pathway to promote osteogenic differentiation.^39,40 This was also demonstrated by KEGG pathway analysis. Notably, miR-690 has been shown to promote osteogenic differentiation by the miR-690/IRS-1/TAZ axis.¹⁵ Hence, it is reasonable to assume that the Wnt signaling pathway or miR-690/IRS-1/TAZ axis may be important pathways in contribution to 3WJ-BMSC_apt/M2-Exos osteogenesis and bone fracture healing.

However, there are still some limitations in this study. First, the study used the mouse macrophage line RAW264.7, rather than primary macrophages, which can be obtained from mouse bone marrow after induction. Macrophages from different sources may have a slight effect on the experimental results. Second, to guide the clinical application of aptamer-functionalized exosomes, we should observe the safety of aptamer-functionalized exosomes after prolonged effects in vivo. Finally, although our results suggested that 3WJ-BMSC_apt/M2-Exos had a targeted therapeutic effect in vivo, to have an in-depth study, further experiments are needed to explore and elucidate the mechanisms by which aptamer-functionalized exosomes promote femur bone fracture healing and interact with cells in the future.

Conclusion

In this study, 3WJ-BMSC_apt/M2-Exos were successfully developed by a novel RNA nanotechnology to improve the targeting ability and aggregation of M2-Exos in the fracture site. This surface modification did not result in the dysfunction of the structure of M2-Exos. Meanwhile, 3WJ-BMSC_apt/M2-Exos can target BMSCs in vitro and exhibit remarked accumulation in the fracture site in vivo, avoiding accumulation in the liver and rapid removal, thus accelerating bone regeneration in a mouse femur fracture model. The targeted delivery of 3WJ-BMSC_apt/M2-Exos provided a novel cell-free therapeutic approach for the treatment of bone fracture. Our study showed that the functionalized exosomes are promising to be used in a targeted drug delivery system for different applications in clinics.

Funding

This work was financially supported by the National Natural Science Foundation of China (Grant number 81671029), the National Major Science and Technology Project of China (Grant number 2016YFC1102900), the Guangzhou Science, Technology and Innovation Commission (Grant number 201803040008), China Scholarship Council (No. 201908440308), and the Talents Introduction of Guangdong Provincial People’s Hospital (Grant number KY0120220255, 3227100558).

Conflict of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Author Contributions

J.S.: collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of the manuscript. S.L.: collection and/or assembly of data, data analysis and interpretation, and final approval of the ­manuscript. Z.Z., W.S., S.Z., Z.G., Z.Y., Z.W., and H.Q.: collection and/or assembly of data, and final approval of the manuscript. J.W.: conception and design, and final approval of the manuscript. M.Z.: conception and design, financial support, and final approval of the manuscript.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.